An energy-efficient scheduling algorithm for shared facility supercomputer centers release_dxyvk7btkrezfoaabnoxd2oxuu

by E.A. Kiselev, P.N. Telegin, B.M. Shabanov

Released as a article .

2021  

Abstract

The evolution of high-performance computing is associated with the growth of energy consumption. Performance of cluster computes (is increased via rising in performance and the number of used processors, GPUs, and coprocessors. An increment in the number of computing elements results in significant growth of energy consumption. Power grids limits for supercomputer centers (SCC) are driving the transition to more energy-efficient solutions. Often upgrade of computing resources is done step-by-step, i.e. parts of older supercomputers are removed from service and replaced with newer ones. A single SCC at any time can operate several computing systems with different performance and power consumption. That is why the problem of scheduling parallel programs execution on SCC resources to optimize energy consumption and minimize the increase in execution time (energy-efficient scheduling) is important. The goal of the presented work was the development of a new energy-efficient algorithm for scheduling computing resources at SCC. To reach the goal the authors analyzed methods of scheduling computing resources in a shared facility, including energy consumption minimizing methods. The study made it possible to formulate the problem of energy-efficient scheduling for a set of CCs and propose an algorithm for its solution. Experiments on NPB benchmarks allowed achieving significant reduction in energy consumption with a minor increase of runtime.
In text/plain format

Archived Files and Locations

application/pdf  713.4 kB
file_4u44qonujbebffeionb2vfm5du
arxiv.org (repository)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article
Stage   submitted
Date   2021-11-17
Version   v1
Language   en ?
arXiv  2111.08978v1
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: ada84fd7-4220-4144-8e0f-ef3b112e2340
API URL: JSON